New brain-computer interface preprint led by PhD student Tyler Singer-Clark! This video of a man with paralysis accurately controlling a cursor looks like something you've seen since ~2017. BUT! This is driven by multielectrode arrays in ventral (speech) motor cortex‼️ 1/

These is the same @BrainGateTeam participant ('T15') and arrays reported in our recent high performance speech neuroprosthesis paper. Except now Tyler added a 2D plus click velocity decoder that enables the participant to use a mouse by attempting hand movements. 2/

Using a rapid online calibration protocol (like what @DrDavidBrandman demonstrated in his 2018 JNE paper), we found that the participant could achieve accurate cursor control in about a minute (training from scratch). 3/

To our surprise, after some hyperparameter optimization, cursor control from the "wrong part of the brain" [is this becoming my catch-phrase🤔?] was the second-best in the published literature (importantly this does not include the recent @neuralink bps records). 4/

We think this is important because it means that with an implant optimized for speech neuroprosthesis performance, good cursor control comes "for free". Now our participant has full keyboard AND mouse control of his personal computer. 🧠➡️⌨️🖱️ Preprint:

very cool!

Thanks Max!

Fantastic work, as always. The flexibility in BCI that this could afford folks with ALS is a potential game changer.

Excellent work, excited to read the pre print. Thanks for sharing!

Amazing, great work!!

Was just looking at the preprint this morning! Super cool! Great work and congrats

Thanks Daniel! You're a bioRxiv early bird eh?